AtomNet PoseRanker: Enriching Ligand Pose Quality for Dynamic Proteins in Virtual High-Throughput Screens

Identification of a selective pyruvate dehydrogenase kinase 1 (PDHK1) chemical probe by virtual screening

PDHK1 is a non-canonical Ser/Thr kinase that negatively regulates the pyruvate dehydrogenase complex (PDC), restricting entry of acetyl-CoA into the tricarboxylic acid (TCA) cycle and downregulating oxidative phosphorylation. In many glycolytic tumors, PDHK1 is overexpressed to suppress activity of the PDC and cause a shift in metabolism toward an increased reliance on glycolysis (the Warburg effect). Genetic studies have shown that knockdown or knockout of PDHK1 reverts this phenotype and inhibits tumor growth in vitro and in vivo, but chemical tools to pharmacologically validate and build upon these data are lacking. We used AtomNet®, a deep convolutional neural network bioactivity predictor, to identify compound 7 as a potential inhibitor of PDHK1. During the process of hit validation, the active species was determined to be isomeric compound 10. Structure-activity studies based on 10 identified 17 as a low μM inhibitor of PDHK1 (IC50 = 1.5 ± 0.3 μM) that is selective against the other PDHK isoforms in both biochemical and cell-based assays. In A549 epithelial lung carcinoma cells, compound 17 inhibits phosphorylation of PDC E1α Ser232, a site that is specifically phosphorylated only by PDHK1, while minimally suppressing phosphorylation of Ser293, a site that is phosphorylated by all four PDHK isoforms. Altogether, these data identify 17 as a selective PDHK1 chemical probe useful for biochemical and cell-based studies.

Development and use of DJ-1 affinity microcolumns to screen and study small drug candidates for Parkinson's disease

BACKGROUND

DJ-1 is a protein whose mutation causes rare heritable forms of Parkinson's disease (PD) and is of interest as a target for treating PD and other disorders. This work used high performance affinity microcolumns to screen and examine the binding of small molecules to DJ-1, as could be used to develop new therapeutics or to study the role of DJ-1 in PD. Non-covalent entrapment was used to place microgram quantities of DJ-1 in an unmodified form within microcolumns, which were then used in multiple studies to analyze binding by model compounds and possible drug candidates to DJ-1.

RESULTS

Several factors were examined in optimizing the entrapment method, including the addition of a reducing agent to maintain a reduced active site cysteine residue in DJ-1, the concentration of DJ-1 employed, and the entrapment times. Isatin was used as a known binding agent (dissociation constant, ∼2.0 μM) and probe for DJ-1 activity. This compound gave good retention on 2.0 cm × 2.1 mm inner diameter DJ-1 microcolumns made under the final entrapment conditions, with a typical retention factor of 14 and elution in ∼8 min at 0.50 mL/min. These DJ-1 microcolumns were used to evaluate the binding of small molecules that were selected in silico to bind or not to bind DJ-1. A compound predicted to have good binding with DJ-1 gave a retention factor of 122, an elution time of ∼15 min at 0.50 mL/min, and an estimated dissociation constant for this protein of 0.5 μM.

SIGNIFICANCE

These chromatographic tools can be used in future work to screen additional possible binding agents for DJ-1 or adapted for examining drug candidates for other proteins. This work represents the first time protein entrapment has been deployed with DJ-1, and it is the first experimental confirmation of binding to DJ-1 by a small lead compound selected in silico.

AI-Based Discovery and CryoEM Structural Elucidation of a KATP Channel Pharmacochaperone

Pancreatic KATP channel trafficking defects underlie congenital hyperinsulinism (CHI) cases unresponsive to the KATP channel opener diazoxide, the mainstay medical therapy for CHI. Current clinically used KATP channel inhibitors have been shown to act as pharmacochaperones and restore surface expression of trafficking mutants; however, their therapeutic utility for KATP trafficking impaired CHI is hindered by high-affinity binding, which limits functional recovery of rescued channels. Recent structural studies of KATP channels employing cryo-electron microscopy (cryoEM) have revealed a promiscuous pocket where several known KATP pharmacochaperones bind. The structural knowledge provides a framework for discovering KATP channel pharmacochaperones with desired reversible inhibitory effects to permit functional recovery of rescued channels. Using an AI-based virtual screening technology AtomNet® followed by functional validation, we identified a novel compound, termed Aekatperone, which exhibits chaperoning effects on KATP channel trafficking mutations. Aekatperone reversibly inhibits KATP channel activity with a half-maximal inhibitory concentration (IC50) ~ 9 μM. Mutant channels rescued to the cell surface by Aekatperone showed functional recovery upon washout of the compound. CryoEM structure of KATP bound to Aekatperone revealed distinct binding features compared to known high affinity inhibitor pharmacochaperones. Our findings unveil a KATP pharmacochaperone enabling functional recovery of rescued channels as a promising therapeutic for CHI caused by KATP trafficking defects.

Current Status and Future Directions of Artificial Intelligence in Post-Traumatic Stress Disorder: A Literature Measurement Analysis

This study aims to explore the current state of research and the applicability of artificial intelligence (AI) at various stages of post-traumatic stress disorder (PTSD), including prevention, diagnosis, treatment, patient self-management, and drug development. We conducted a bibliometric analysis using software tools such as Bibliometrix (version 4.1), VOSviewer (version 1.6.19), and CiteSpace (version 6.3.R1) on the relevant literature from the Web of Science Core Collection (WoSCC). The analysis reveals a significant increase in publications since 2017. Kerry J. Ressler has emerged as the most influential author in the field to date. The United States leads in the number of publications, producing seven times more papers than Canada, the second-ranked country, and demonstrating substantial influence. Harvard University and the Veterans Health Administration are also key institutions in this field. The Journal of Affective Disorders has the highest number of publications and impact in this area. In recent years, keywords related to functional connectivity, risk factors, and algorithm development have gained prominence. The field holds immense research potential, with AI poised to revolutionize PTSD management through early symptom detection, personalized treatment plans, and continuous patient monitoring. However, there are numerous challenges, and fully realizing AI's potential will require overcoming hurdles in algorithm design, data integration, and societal ethics. To promote more extensive and in-depth future research, it is crucial to prioritize the development of standardized protocols for AI implementation, foster interdisciplinary collaboration-especially between AI and neuroscience-and address public concerns about AI's role in healthcare to enhance its acceptance and effectiveness.

Discovery of a Small Molecule with an Inhibitory Role for RAB11

RAB11, a pivotal RabGTPase, regulates essential cellular processes such as endocytic recycling, exocytosis, and autophagy. The protein was implicated in various human diseases, including cancer, neurodegenerative disorders, viral infections, and podocytopathies. However, a small-molecular inhibitor is lacking. The complexity and workload associated with potential assays make conducting large-scale screening for RAB11 challenging. We employed a tiered approach for drug discovery, utilizing deep learning-based computational screening to preselect compounds targeting a specific pocket of RAB11 protein with experimental validation by an in vitro platform reflecting RAB11 activity through the exocytosis of GFP. Further validation included the exposure of Drosophila by drug feeding. In silico pre-screening identified 94 candidates, of which 9 were confirmed using our in vitro platform for Rab11 activity. Focusing on compounds with high potency, we assessed autophagy, which independently requires RAB11, and validated three of these compounds. We further analyzed the dose-response relationship, observing a biphasic, potentially hormetic effect. Two candidate compounds specifically caused a shift in Rab11 vesicles to the cell periphery, without significant impact on Rab5 or Rab7. Drosophila larvae exposed to another candidate compound with predicted oral bioavailability exhibited minimal toxicity, subcellular dispersal of endogenous Rab11, and a decrease in RAB11-dependent nephrocyte function, further supporting an inhibitory role. Taken together, the combination of computational screening and experimental validation allowed the identification of small molecules that modify the function of Rab11. This discovery may further open avenues for treating RAB11-associated disorders.

Addressing docking pose selection with structure-based deep learning: Recent advances, challenges and opportunities

Molecular docking is a widely used technique in drug discovery to predict the binding mode of a given ligand to its target. However, the identification of the near-native binding pose in docking experiments still represents a challenging task as the scoring functions currently employed by docking programs are parametrized to predict the binding affinity, and, therefore, they often fail to correctly identify the ligand native binding conformation. Selecting the correct binding mode is crucial to obtaining meaningful results and to conveniently optimizing new hit compounds. Deep learning (DL) algorithms have been an area of a growing interest in this sense for their capability to extract the relevant information directly from the protein-ligand structure. Our review aims to present the recent advances regarding the development of DL-based pose selection approaches, discussing limitations and possible future directions. Moreover, a comparison between the performances of some classical scoring functions and DL-based methods concerning their ability to select the correct binding mode is reported. In this regard, two novel DL-based pose selectors developed by us are presented.

HDBind: encoding of molecular structure with hyperdimensional binary representations

Traditional methods for identifying "hit" molecules from a large collection of potential drug-like candidates rely on biophysical theory to compute approximations to the Gibbs free energy of the binding interaction between the drug and its protein target. These approaches have a significant limitation in that they require exceptional computing capabilities for even relatively small collections of molecules. Increasingly large and complex state-of-the-art deep learning approaches have gained popularity with the promise to improve the productivity of drug design, notorious for its numerous failures. However, as deep learning models increase in their size and complexity, their acceleration at the hardware level becomes more challenging. Hyperdimensional Computing (HDC) has recently gained attention in the computer hardware community due to its algorithmic simplicity relative to deep learning approaches. The HDC learning paradigm, which represents data with high-dimension binary vectors, allows the use of low-precision binary vector arithmetic to create models of the data that can be learned without the need for the gradient-based optimization required in many conventional machine learning and deep learning methods. This algorithmic simplicity allows for acceleration in hardware that has been previously demonstrated in a range of application areas (computer vision, bioinformatics, mass spectrometery, remote sensing, edge devices, etc.). To the best of our knowledge, our work is the first to consider HDC for the task of fast and efficient screening of modern drug-like compound libraries. We also propose the first HDC graph-based encoding methods for molecular data, demonstrating consistent and substantial improvement over previous work. We compare our approaches to alternative approaches on the well-studied MoleculeNet dataset and the recently proposed LIT-PCBA dataset derived from high quality PubChem assays. We demonstrate our methods on multiple target hardware platforms, including Graphics Processing Units (GPUs) and Field Programmable Gate Arrays (FPGAs), showing at least an order of magnitude improvement in energy efficiency versus even our smallest neural network baseline model with a single hidden layer. Our work thus motivates further investigation into molecular representation learning to develop ultra-efficient pre-screening tools. We make our code publicly available at https://github.com/LLNL/hdbind .

NGT: Generative AI with Synthesizability Guarantees Discovers MC2R Inhibitors from a Tera-Scale Virtual Screen

Commercially available, synthesis-on-demand virtual libraries contain upward of trillions of readily synthesizable compounds for drug discovery campaigns. These libraries are a critical resource for rapid cycles of in silico discovery, property optimization and in vitro validation. However, as these libraries continue to grow exponentially in size, traditional search strategies encounter significant limitations. Here we present NeuralGenThesis (NGT), an efficient reinforcement learning approach to generate compounds from ultralarge libraries that satisfy user-specified constraints. Our method first trains a generative model over a virtual library and subsequently trains a normalizing flow to learn a distribution over latent space that decodes constraint-satisfying compounds. NGT allows multiple constraints simultaneously without dictating how molecular properties are calculated. Using NGT, we generated potent and selective inhibitors for the melanocortin-2 receptor (MC2R) from a three trillion compound library. NGT offers a powerful and scalable solution for navigating ultralarge virtual libraries, accelerating drug discovery efforts.

Artificial Intelligence (AI) Applications in Drug Discovery and Drug Delivery: Revolutionizing Personalized Medicine

Artificial intelligence (AI) encompasses a broad spectrum of techniques that have been utilized by pharmaceutical companies for decades, including machine learning, deep learning, and other advanced computational methods. These innovations have unlocked unprecedented opportunities for the acceleration of drug discovery and delivery, the optimization of treatment regimens, and the improvement of patient outcomes. AI is swiftly transforming the pharmaceutical industry, revolutionizing everything from drug development and discovery to personalized medicine, including target identification and validation, selection of excipients, prediction of the synthetic route, supply chain optimization, monitoring during continuous manufacturing processes, or predictive maintenance, among others. While the integration of AI promises to enhance efficiency, reduce costs, and improve both medicines and patient health, it also raises important questions from a regulatory point of view. In this review article, we will present a comprehensive overview of AI's applications in the pharmaceutical industry, covering areas such as drug discovery, target optimization, personalized medicine, drug safety, and more. By analyzing current research trends and case studies, we aim to shed light on AI's transformative impact on the pharmaceutical industry and its broader implications for healthcare.

Antimicrobial activity of compounds identified by artificial intelligence discovery engine targeting enzymes involved in Neisseria gonorrhoeae peptidoglycan metabolism

BACKGROUND

Neisseria gonorrhoeae (Ng) causes the sexually transmitted disease gonorrhoea. There are no vaccines and infections are treated principally with antibiotics. However, gonococci rapidly develop resistance to every antibiotic class used and there is a need for developing new antimicrobial treatments. In this study we focused on two gonococcal enzymes as potential antimicrobial targets, namely the serine protease L,D-carboxypeptidase LdcA (NgO1274/NEIS1546) and the lytic transglycosylase LtgD (NgO0626/NEIS1212). To identify compounds that could interact with these enzymes as potential antimicrobials, we used the AtomNet virtual high-throughput screening technology. We then did a computational modelling study to examine the interactions of the most bioactive compounds with their target enzymes. The identified compounds were tested against gonococci to determine minimum inhibitory and bactericidal concentrations (MIC/MBC), specificity, and compound toxicity in vitro.

RESULTS

AtomNet identified 74 compounds that could potentially interact with Ng-LdcA and 84 compounds that could potentially interact with Ng-LtgD. Through MIC and MBC assays, we selected the three best performing compounds for both enzymes. Compound 16 was the most active against Ng-LdcA, with a MIC50 value < 1.56 µM and MBC50/90 values between 0.195 and 0.39 µM. In general, the Ng-LdcA compounds showed higher activity than the compounds directed against Ng-LtgD, of which compound 45 had MIC50 values of 1.56-3.125 µM and MBC50/90 values between 3.125 and 6.25 µM. The compounds were specific for gonococci and did not kill other bacteria. They were also non-toxic for human conjunctival epithelial cells as judged by a resazurin assay. To support our biological data, in-depth computational modelling study detailed the interactions of the compounds with their target enzymes. Protein models were generated in silico and validated, the active binding sites and amino acids involved elucidated, and the interactions of the compounds interacting with the enzymes visualised through molecular docking and Molecular Dynamics Simulations for 50 ns and Molecular Mechanics Poisson-Boltzmann Surface Area (MM-PBSA).

CONCLUSIONS

We have identified bioactive compounds that appear to target the N. gonorrhoeae LdcA and LtgD enzymes. By using a reductionist approach involving biological and computational data, we propose that compound Ng-LdcA-16 and Ng-LtgD-45 are promising anti-gonococcal compounds for further development.

Identification of potent schistosomicidal compounds predicted as type II-kinase inhibitors against Schistosoma mansoni c-Jun N-terminal kinase SMJNK

Introduction

Schistosomiasis has for many years relied on a single drug, praziquantel (PZQ) for treatment of the disease. Immense efforts have been invested in the discovery of protein kinase (PK) inhibitors; however, given that the majority of PKs are still not targeted by an inhibitor with a useful level of selectivity, there is a compelling need to expand the chemical space available for synthesizing new, potent, and selective PK inhibitors. Small-molecule inhibitors targeting the ATP pocket of the catalytic domain of PKs have the potential to become drugs devoid of (major) side effects, particularly if they bind selectively. This is the case for type II PK inhibitors, which cause PKs to adopt the so-called DFG-out conformation, corresponding to the inactive state of the enzyme.

Methods

The goal was to perform a virtual screen against the ATP pocket of the inactive JNK protein kinase. After virtually screening millions of compounds, Atomwise provided 85 compounds predicted to target c-Jun N-terminal kinase (JNK) as type II inhibitors. Selected compounds were screened in vitro against larval stage (schistosomula) of S. mansoni using the XTT assay. Adult worms were assessed for motility, attachment, and pairing stability. Active compounds were further analyzed by molecular docking against SmJNK.

Results

In total, 33 compounds were considered active in at least one of the assays, and two compounds were active in every in vitro screening assay. The two most potent compounds presented strong effects against both life stages of the parasite, and microscopy analysis showed phenotypic alterations on the tegument, in the gonads, and impairment of cell proliferation.

Conclusion

The approach to screen type II kinase inhibitors resulted in the identification of active compounds that will be further developed against schistosomiasis.

AI is a viable alternative to high throughput screening: a 318-target study

High throughput screening (HTS) is routinely used to identify bioactive small molecules. This requires physical compounds, which limits coverage of accessible chemical space. Computational approaches combined with vast on-demand chemical libraries can access far greater chemical space, provided that the predictive accuracy is sufficient to identify useful molecules. Through the largest and most diverse virtual HTS campaign reported to date, comprising 318 individual projects, we demonstrate that our AtomNet® convolutional neural network successfully finds novel hits across every major therapeutic area and protein class. We address historical limitations of computational screening by demonstrating success for target proteins without known binders, high-quality X-ray crystal structures, or manual cherry-picking of compounds. We show that the molecules selected by the AtomNet® model are novel drug-like scaffolds rather than minor modifications to known bioactive compounds. Our empirical results suggest that computational methods can substantially replace HTS as the first step of small-molecule drug discovery.

A generalized protein-ligand scoring framework with balanced scoring, docking, ranking and screening powers

Applying machine learning algorithms to protein-ligand scoring functions has aroused widespread attention in recent years due to the high predictive accuracy and affordable computational cost. Nevertheless, most machine learning-based scoring functions are only applicable to a specific task, e.g., binding affinity prediction, binding pose prediction or virtual screening, suggesting that the development of a scoring function with balanced performance in all critical tasks remains a grand challenge. To this end, we propose a novel parameterization strategy by introducing an adjustable binding affinity term that represents the correlation between the predicted outcomes and experimental data into the training of mixture density network. The resulting residue-atom distance likelihood potential not only retains the superior docking and screening power over all the other state-of-the-art approaches, but also achieves a remarkable improvement in scoring and ranking performance. We emphatically explore the impacts of several key elements on prediction accuracy as well as the task preference, and demonstrate that the performance of scoring/ranking and docking/screening tasks of a certain model could be well balanced through an appropriate manner. Overall, our study highlights the potential utility of our innovative parameterization strategy as well as the resulting scoring framework in future structure-based drug design.

Artificial Intelligence in Pharmaceutical Technology and Drug Delivery Design

Artificial intelligence (AI) has emerged as a powerful tool that harnesses anthropomorphic knowledge and provides expedited solutions to complex challenges. Remarkable advancements in AI technology and machine learning present a transformative opportunity in the drug discovery, formulation, and testing of pharmaceutical dosage forms. By utilizing AI algorithms that analyze extensive biological data, including genomics and proteomics, researchers can identify disease-associated targets and predict their interactions with potential drug candidates. This enables a more efficient and targeted approach to drug discovery, thereby increasing the likelihood of successful drug approvals. Furthermore, AI can contribute to reducing development costs by optimizing research and development processes. Machine learning algorithms assist in experimental design and can predict the pharmacokinetics and toxicity of drug candidates. This capability enables the prioritization and optimization of lead compounds, reducing the need for extensive and costly animal testing. Personalized medicine approaches can be facilitated through AI algorithms that analyze real-world patient data, leading to more effective treatment outcomes and improved patient adherence. This comprehensive review explores the wide-ranging applications of AI in drug discovery, drug delivery dosage form designs, process optimization, testing, and pharmacokinetics/pharmacodynamics (PK/PD) studies. This review provides an overview of various AI-based approaches utilized in pharmaceutical technology, highlighting their benefits and drawbacks. Nevertheless, the continued investment in and exploration of AI in the pharmaceutical industry offer exciting prospects for enhancing drug development processes and patient care.

Discovery of a cryptic pocket in the AI-predicted structure of PPM1D phosphatase explains the binding site and potency of its allosteric inhibitors

Virtual screening is a widely used tool for drug discovery, but its predictive power can vary dramatically depending on how much structural data is available. In the best case, crystal structures of a ligand-bound protein can help find more potent ligands. However, virtual screens tend to be less predictive when only ligand-free crystal structures are available, and even less predictive if a homology model or other predicted structure must be used. Here, we explore the possibility that this situation can be improved by better accounting for protein dynamics, as simulations started from a single structure have a reasonable chance of sampling nearby structures that are more compatible with ligand binding. As a specific example, we consider the cancer drug target PPM1D/Wip1 phosphatase, a protein that lacks crystal structures. High-throughput screens have led to the discovery of several allosteric inhibitors of PPM1D, but their binding mode remains unknown. To enable further drug discovery efforts, we assessed the predictive power of an AlphaFold-predicted structure of PPM1D and a Markov state model (MSM) built from molecular dynamics simulations initiated from that structure. Our simulations reveal a cryptic pocket at the interface between two important structural elements, the flap and hinge regions. Using deep learning to predict the pose quality of each docked compound for the active site and cryptic pocket suggests that the inhibitors strongly prefer binding to the cryptic pocket, consistent with their allosteric effect. The predicted affinities for the dynamically uncovered cryptic pocket also recapitulate the relative potencies of the compounds (τ _b =0.70) better than the predicted affinities for the static AlphaFold-predicted structure (τ _b =0.42). Taken together, these results suggest that targeting the cryptic pocket is a good strategy for drugging PPM1D and, more generally, that conformations selected from simulation can improve virtual screening when limited structural data is available.

Comprehensive Survey of Consensus Docking for High-Throughput Virtual Screening

The rapid advances of 3D techniques for the structural determination of proteins and the development of numerous computational methods and strategies have led to identifying highly active compounds in computer drug design. Molecular docking is a method widely used in high-throughput virtual screening campaigns to filter potential ligands targeted to proteins. A great variety of docking programs are currently available, which differ in the algorithms and approaches used to predict the binding mode and the affinity of the ligand. All programs heavily rely on scoring functions to accurately predict ligand binding affinity, and despite differences in performance, none of these docking programs is preferable to the others. To overcome this problem, consensus scoring methods improve the outcome of virtual screening by averaging the rank or score of individual molecules obtained from different docking programs. The successful application of consensus docking in high-throughput virtual screening highlights the need to optimize the predictive power of molecular docking methods.

Protein-Ligand Docking in the Machine-Learning Era

Molecular docking plays a significant role in early-stage drug discovery, from structure-based virtual screening (VS) to hit-to-lead optimization, and its capability and predictive power is critically dependent on the protein-ligand scoring function. In this review, we give a broad overview of recent scoring function development, as well as the docking-based applications in drug discovery. We outline the strategies and resources available for structure-based VS and discuss the assessment and development of classical and machine learning protein-ligand scoring functions. In particular, we highlight the recent progress of machine learning scoring function ranging from descriptor-based models to deep learning approaches. We also discuss the general workflow and docking protocols of structure-based VS, such as structure preparation, binding site detection, docking strategies, and post-docking filter/re-scoring, as well as a case study on the large-scale docking-based VS test on the LIT-PCBA data set.

Scoring Functions for Protein-Ligand Binding Affinity Prediction using Structure-Based Deep Learning: A Review

The rapid and accurate in silico prediction of protein-ligand binding free energies or binding affinities has the potential to transform drug discovery. In recent years, there has been a rapid growth of interest in deep learning methods for the prediction of protein-ligand binding affinities based on the structural information of protein-ligand complexes. These structure-based scoring functions often obtain better results than classical scoring functions when applied within their applicability domain. Here we review structure-based scoring functions for binding affinity prediction based on deep learning, focussing on different types of architectures, featurization strategies, data sets, methods for training and evaluation, and the role of explainable artificial intelligence in building useful models for real drug-discovery applications.